We have assembled a team of multi-disciplinary investigators to study, on a population basis, the patterns of treatment initially and continuing for a representative sample of lung cancer patients in the state of Iowa. These patterns will be explored in terms of patient decision-making processes for short and long-term treatment, with follow-up for suitable outcomes including disease-free interval, quality-of-life measures, end of-life care and overall survivorship. Factors proposed for exploration include including patient demographic, economic and behavioral attributes, including traits such as the patient's cancer learning environment and fatalism, the influences of geographically-dispersed practitioners with varied clinical management approaches, practitioner continuing education and information sources, the application of treatments where optimal approaches are uncertain and the influence of spatial and fiscal access to care. Special features of this proposal include: a) a multi-disciplinary team of investigators including epidemiology, disease registration, biostatistics, health economics, behavioral science and thoracic oncology; b) proven ability to access approximately 2000 lung cancer patients per year and achieve long term follow-up rates of 98-99%, using the Iowa Cancer Registry (NCI SEER Program); c) the elaboration of an economics model to guide both theory and questionnaire design for patients and practitioners, and offered to understand patient decision-making and its relation to outcomes; and d) demonstrable high levels of community cooperation, from patients, cancer care practitioners and pathologists. The goal of the special study is to use Instrumental Variable (IV) techniques to assess whether patient survival rates can be increased by increasing adjuvant therapy treatment rates for early stage non-small cell lung cancer (ENSCLC) patients. If providers are accurately sorting patients by their expected treatment benefits, and only patients expected to benefit from adjuvant treatments actually receive them, we would expect that treatment benefits for patients on the extensive margin to be minimal. We hypothesize there will be little or no treatment survival benefits for patients on the ENSCLC patients provides the rationale for this study.